Dauer: 1:51:03
Alice Zhao
pyohio.org/2018/schedule/presentation/38/
Natural language processing (NLP) is an exciting branch of artificial intelligence (AI) that allows machines to break down and understand human language. As a data scientist, I often use NLP techniques to interpret text data that I'm working with for my analysis. During this tutorial, I plan to walk through text pre-processing techniques, machine learning techniques and Python libraries for NLP.
Text pre-processing techniques include tokenization, text normalization and data cleaning. Once in a standard format, various machine learning techniques can be applied to better understand the data. This includes using popular modeling techniques to classify emails as spam or not, or to score the sentiment of a tweet on Twitter. Newer, more complex techniques can also be used such as topic modeling, word embeddings or text generation with deep learning.
We will walk through an example in Jupyter Notebook that goes through all of the steps of a text analysis project, using several NLP libraries in Python including NLTK, TextBlob, spaCy and gensim along with the standard machine learning libraries including pandas and scikit-learn.
## Setup Instructions
[ github.com/adashofdata/nlp-in-python-tutorial](github.com/adashofdata/nlp-in-python-tutorial)
===
pyohio.org
A FREE annual conference for anyone interested in Python in and around Ohio, the entire Midwest, maybe even the whole world.
Aufrufe 355 302
Aufrufe 17 000
Aufrufe 418 000
Aufrufe 1 640 453
KOMMENTARE
Christopher R Seay
this really shows why knowing your data matters. louis ck's acts are generally depressing, not positive, but the textblob analysis doesn't catch sarcasm. mike's acts are also generally not upbeat.
Vor 9 TageKhan
Hey, time stamps for the examples?
Vor 15 TageLev Che
This lecture would be better if authors transcribed questions cause they are totally lost.
Vor 17 TagePouya Naseri
excellent : Sentiment(polarity=1.0, subjectivity=1.0) 😉
Vor 23 Tagehassan jan
This tutorial was super, Thank's alot. any one want to help me in research proposal.
Vor 23 TageLevente Hegyi
Amazing!
Vor 23 TageDeepesh Giri
I don't know if anyone faced this issue but "(class_ = "post-content")" @ 35:00 doesn't work anymore. The scrapsfromtheloft.com website now shows "(class_ = "ast-container")" and making this change worked for me.
Vor 24 TageMalcolm Ng
Thank you Alice: I've learned more about how to think like a data science in this one video than I ever did on the job
Vor 27 Tagesirshendu2e01
Agreed .. I've been watching a couple of NLP tutorials for about a week and this one is the most satisfying one yet
Vor 17 TageYongjae Park
This contents truly helped me!
Vor MonatDaniel Marx
Is it better to use pickle vs just saving df's into csv files to be read back later? Why or why not?
Vor MonatDiego Molina
She is an amazing teacher, what a great lecture ;)
Vor MonatMELS E-Learn Dot Com
good
Vor Monateswari jayakumar
Excellent video ! She is a great teacher !!!
Vor MonatD Mukherjee
Interesting and Useful material presented lucidly. Thanks a lot Alice
Vor MonatAhmad Zaim Hilmi
Found this in 2021. Great lecture!!
Vor MonatMuhammad Yasir
Very good lecture. Can you send me the link of slides use in this tutorial.
Vor MonatMuhammad Yasir
1:08:02
Vor MonatDarcy Dixon
Beautiful love u.
Vor MonatMd. Niamul Ahad Chowdhury
Amid my relaxation in a seabeach of tranquility, a cough wave came and washed my sandcastle away.
Vor 2 MonateMd. Niamul Ahad Chowdhury
It would be helpful if there was some idea of how long these processes take—writing the code, then running the programs. I imagine it takes quite a while to spend time on each word, e.g. in LDA.
Vor 2 MonateMd. Niamul Ahad Chowdhury
This session made me realise NLP might be my niche. Also, what a great instructor to have introduce it to me! 💛💛💛
Vor 2 MonateWorkneh Chalchisa
Now I know what I'm going to do on my research.... Thank you enlighten me
Vor 2 MonateSaed Sayad
Excellent!
Vor 2 MonatePrateek Bhardwaj
Can anybody tell me if there's a change in the code at Git, I am not able to work on it.
Vor 2 MonateNikhil Rana
naah, its working just fine. I had some glitches tho. working on it.
Vor 2 MonateManuel Herrera Hipnotista
One of the best presentions I have seen in my lifetime! (and I have seen a lot) Clear, simple and to the point! Thanks
Vor 2 MonateAli Akbar
How can I get the presentation file?
Vor 2 MonateViviana Ventura
Very useful and you make seem all simple!
Vor 2 MonateSrijita Saha Roy
How to clean data for pdf files?
Vor 2 MonateVuc Tuc
Poor dog I liked my own commet how sad.
Vor 2 MonateSelf Learning
Could i get the code? my email is :emrul.phy@gmail.com. Thanks
Vor 2 MonateSelf Learning
Best resource i have been studying Data science. Just loved it! Thanks for the nice presentation
Vor 2 MonateBrenden Song
Great class! Thank you!
Vor 2 MonateHu Huy
MY SЕХ LOVE.
Vor 2 MonateLeonardo Galindo
OMG! By far, the best 2 hrs that I've spend in YouTube in the last years... On behalf of this Data Scientist "in progress", Thank you so much for posting this content.
Vor 2 MonateSowmik Sarker
Excellent teacher with deliberate and relaxed teaching style!!! and wish if every teachers be like you. Thanks for giving such a tidy lecture.
Vor 3 MonateShervin Maleki
Very useful video for NLP learners! Highly recommended!
Vor 3 MonateEustaquio Santimano
Easy to follow training !!
Vor 3 MonateADITYA RAJ
Thanks for this wonderful lecture.
Vor 3 MonateEvry
This is gold, thank you a lot!
Vor 3 MonateAmila Priyanga
https://www.youtube.com/watch?v=v_U8h2sNIDw
Vor 3 Monate78 DA
This is so cool, and it is very helpful.
Vor 3 MonateLin Qin
Hi, May I find the slides somewhere?
Vor 3 MonateHimanshu Bag
just amazing ... this 2 hr video felt like a 20 min video .... so much concept , well detailed ... totally worth it
Vor 3 MonateVaibhav Patil
Top explination ever seen!!
Vor 4 MonateSteve A-G
Wow! This was incredible. Thanks for doing this in an approachable and concise way.
Vor 4 MonateThe Zodiac
Alice is so pleasant to listen to. Enjoyed this presentation a lot. super helpful.
Vor 4 MonateVishal Pulikottil
You're a brilliant teacher. So, so helpful.
Vor 4 Monateguy fishman
Im in the danger zone 🤷🏼
Vor 4 MonateWahranRai
Chinese everywhere ! where is TRUMP ?
Vor 5 MonateFOR Vpbof-20
Ms. Alice, after watching this lecture in my home office I stood up and applauded as if I watched a piece of art or an amazing movie. Thank you.
Vor 5 Monateshalu chaudhary
In the step load pickle file. It is showing error2- no such file or directory. I have used the exact data she is using than also it's giving this error y so. Please do let me if anyone know.
Vor 6 MonateKarthik Ks
Hi... Thank you for your wonderful session. I am new to NLP. I have a dataset, which has 5000 records/samples with targets. For every (or less than)10 records, there is one Target/class. In other words, there are 500 targets/classes approximately for 5000 samples. Besides, each target/class is multiple words (like a simple sentence). I really seek your advise/help in building a (NLP) model using the data set with this kind of target. Thank you.
Vor 6 MonateIndu Pandey
https://youtu.be/EEYwJAlMoTc
Vor 6 Monatehateclub
All the people after Topic Modelling when she said "I'm running out of time", who then proceeded to waste her time by asking a ton of questions, are the worst kind of people in lectures.
Vor 6 MonateDon Eldorado
Super good!
Vor 6 MonateA4 Analysis
very good RE for any programming language i used it during PERL programming
Vor 6 Monate13 eau
love the comedian examples. really easy to understand.
Vor 6 MonateLarry Orlando
Don’t give up so quick, online trading has been working out for me. I decided to share this because I want others to benefit from this opportunity, I strongly recommend this manager to those that look for a stable system..You can contact him via Instagram page @fx_with_jake_dollery
Vor 6 Monatemahmoud abdelsattar
very nice thanks a lot keep it up.
Vor 6 MonateDhyanaja Alva A
Throw him out of the class
Vor 7 Monatefx ea
https://www.youtube.com/watch?v=LqpgGfCv3d8
Vor 7 MonateGood to know
All coding aside Ricky Gervais unquestionably is the best comedian amongst all
Vor 7 Monatebilboswaggins76
This video is amazing! I really, really like how she goes through the cleansing process because data is almost never cookie cutter like we want it to be. Loved seeing all the steps laid out so methodically.
Vor 7 Monatemohamed meerasa
Thanks a lot
Vor 7 Monatenagesh sutar
Thor
Vor 7 Monatebasu ml
The trainer is really awesome, very knowledgeable I blessed to check your class Thank you, Alice, for such an incredible session
Vor 7 MonateJesús Valenzuela
Excelente presentacion.
Vor 7 MonateRandom Person
audience members who ask questions during a talk should be bullied into silence
Vor 7 MonateDivya Choudhary
Great Speaker and Explanation...would love to learn more from you.
Vor 7 Monate안해찬
This is better than one semester NLP course I took.
Vor 7 MonateJavier Gamonal
there's a lot of good courses on the internet that are on the level of a college course!
Vor 13 Tageopensource Reproduction
I Have A Idea , Maybe A Dumb , But every conversation depends on previous topic or previous talk , so we need to compare current talk with previous to get more accurate results ?? Dumb Or Good?
Vor 8 MonateFarhat Ahmed
Very clear lesson. Incredibly useful. I can’t thank you enough. Really saw the power of data science using the stuff in this tutorial. Thank you!!!
Vor 8 MonateKhaki
Thank you very useful.
Vor 8 MonateAflouat Abdel Wedoud
thanks
Vor 8 MonateMandeep Singh
9:56 yeeeehhhhhhhhh!!!!
Vor 8 MonatePedro Neves
idk why i just watched this . i started watching vids to study for my java exam and ended up here . congrats on this presentation it was really interesting and insightful
Vor 8 MonateNandini nans
Brilliant Tutorial, tnx for sharing. Any updates to consider since this recording ? Is it still valid ? Kindly suggest.
Vor 8 MonateRadha Jayaraman
one of the best videos !!! so insightful
Vor 8 MonateHakim
As an Asian your life motto meant a lot to me, thank you for reminding me to let go of perfectionism because she doesn't exists :)
Vor 8 MonateGraham Nicholls
Great speaker - enthusiastic, clear, cogent, engaging, and above all interesting. Superb.
Vor 8 MonateMohit Nagarkoti
The Best Video to get you started with NLP
Vor 8 MonateJana Hokszová
It's been a pleasure learning from this lecture. I am trying to figure out how to work with large text files (judgments) for my thesis project and wish I had more guidance like this along the way as a beginner in the area, most of the material I found before this starts with a corpus and I'm there trying to figure out what exactly that is and how to make my dataset into it. Thank you!
Vor 8 MonateJonathan Chow
This was amazing and helpful. thanks.
Vor 8 MonateRalph MacFadden
By removing punctuation and capitals, you sometimes change the meaning of the text. When will we eat, Honey? when will we eat honey
Vor 8 Monateyudhi riyansyah
Hi, can you share the slide of presentation ? Thanks
Vor 8 MonateAb Ed
1:02:12 Bo Burnham with "repeat" and "stuff" in the top words and next to each other. I can die in peace!
Vor 8 MonateHugo Rojas
Very Good!
Vor 8 MonateTheophilous Mathema
Great tutorial, l really enjoyed it. thanks for sharing
Vor 8 MonateBhavya Parikh
Very Great Tutorail to strart NLP good explanation
Vor 8 MonateMarkus Koy
Very good tutorial. I learned a lot. Thanks so much.
Vor 8 MonateCosmic Burrito
8:50 I didnt come here to be attacked
Vor 8 MonateLuisa Ferrari
This is an amazing and really helpful tutorial, thank you!
Vor 8 MonateSatwik Ch
I don't think of doing sentiment analysis on the comment section of this video. I don't think I can find any negative reviews :) Thank you, Alice!!
Vor 8 MonateAllSeriousStuff(ASS)
Hello Alice, please cover other AI ML topics too as I like most of your explanation and examples you present are unique.
Vor 8 MonateMarsLanding91
Alice - this was fantastic, thank you!
Vor 8 Monatekui0926
clap clap clap
Vor 8 MonateRahul Pathak
Is the code available ?
Vor 8 Monatedragonsder
I'm a bit confused by her use of regular expressions around 41:00. The 1st and 3rd expressions make sense, but the second ('[%s] % re.escape(string.punctuation), '', text) is a bit confusing. It looks like %s is a scanf() token, not a regular expression and I'm not sure why she switches between regex and scanf() format. Can anyone explain why this is advantageous?
Vor 8 MonatePaul Ulrich
Nice talk, but in the exploratory data analysis (at 1:03), to determine the size of each comedian’s vocabulary, the presenter Alice Zhao should have normalized the comparison by dividing the number of each comedian’s unique words by the total words in the corresponding stand-up routine. In doing so, she would have seen that her favorite Ali Wong had the highest percentage at 41%, followed closely by Bo Burnham at 40%. All the others were at or below 33%. In other words, based on their vocabulary, Ali and Bo are probably the most literate (intelligent) of the comics. Also Ali’s name is similar to Alice, another factor that may explain the speaker’s affinity to her.
Vor 8 MonateAndrew Wright
Alie Zhao - great tutorial. Thanks
Vor 8 Monate